Demand Planning Pain Points

There are dozens of Demand Planning Pain Points and each of them can have multiple irritations. From Information Technology to Data and from Resource to Process. Many of these aches are not exclusive to Demand Planning but can be felt all across business activities. The most common and painful of these challenges are discussed below.

The Panaceas to these problems are often sought in System Implementations or upgrades (Shiny New Tool!) or in Process invigoration (Sales & Operation Planning or Integrated Business Planning) and when performed well, these changes can transform forecasting capabilities. Done without proper assessment and strategy though, these changes just provide new Pitfalls to fall into.

Information Technology

Painfully Slow

Is the Statistical Engine run taking a long time? What happens if it fails during the weekend (perhaps the only time available)? Do you run it again during the week, which means nobody can plan until it is finished? Does it take multiple minutes to open a spreadsheet? How long to export the full dataset? Is that even possible?

There are two absolutely essential tasks that Demand Planners need to do with their data: Firstly, it is to be as accurate as possible. Secondly, it is to be as fast as possible. Fast in this regard means hitting the cycle time points but also to be able to react really quickly at any time and re-forecast. A slow Demand Planning System is extremely painful for your Planners and your Business.

Licensing to kill

The current costs are probably not compatible with system capability. The world is rapidly evolving, and Demand Planning should not remain static. Is the software support running out? Being forced to upgrade? Will there be extended support costs? Perhaps you are restricted to a certain number of users or prevented from utilising all the functionality that you would like. BlockChain, Machine Learning and the Internet of Things are not just buzzwords - they will be transformative.

Hurtful Hardware

On-premise software requires hardware too. Sometimes lots of it, plus backups. And IT resources to run them. The result will be multiple servers for a system that is potentially not returning value. This is not a Demand Planning issue alone though; this is a headache everywhere.

Injurious Integration

Old systems require old file formats & prevent change in source systems and IT methodology too. I expect you are collecting data from source systems, then converting into .CSV, then checking for errors & manually loading (probably in chunks). The same steps happen at the end of the Demand Planning process: exporting the data into Excel so that it can be converted into a format for loading into another spreadsheet or system.

Sensitive Security

Can your planners see all the data or are they restricted? Demand Planning solutions can contain extremely sensitive data - especially if Revenue, Cost and Margin are involved. All your history and all your future, for every finished product, component and spare. That's a lot of data you don't want just lying around. If the all planners share a full dataset - can you audit who has changed what, when, by how much and why? How easy is it to get access?

The problem exists if your solution is Excel based too. Here, access is probably restricted to only one user at a time, but the data can go anywhere at any time. Or perhaps you have multiple Excel Plans in operation at the same time? That is an excruciating thing to control and consolidate. Formulas, Lookups, Imports... all this insecurity is an accident waiting to happen.

Miserable Maintenance & Rotten Risk

If you are running old systems or (Excel solutions) what if the system (or the laptop it is on) crashes? If your cycle time is in Months and you get a failure just before the upload file is to be sent, and everything is lost... Yelp! Can you easily use the last cycle data while waiting for support and recovery? Do you even have backups of all your old files and forecasts for access and comparison?

Olde World Functionality

At the most basic level, Demand Planners need two types of User Interface functionality. They need large and flexible tables, and they need impactful graphs. The graphs should cover multiple styles from linear to pie and everything in-between. Demand Planning Managers and Business Executives need Dashboards that present tables and graphs at aggregated hierarchy levels.

Beyond these truly basic front-end functions, if you can't filter, exception, sort, drill, expand, update, re-run, comment and save in an efficient manner then your functionality is old. If you cannot manage data by Categories, Groups, Attributes, Classification, Life-Cycle, Criticality or Accuracy then your functionality is old.

If you cannot access multiple data streams, apply on-the-fly calculations, review casual, syndicated and IoT data, then your functionality is old. If you are unable to automate statistical engine simulations with machine learning, manage new product introductions, workflow and review life-cycle changes, access promotion calendars and inventory overstock warnings, then your functionality is old.

If your planners cannot review and update their data on a mobile device, then your functionality is old. If you cannot segment your data into specific plans by Channel, Territory, Organisation, Manufacturing Plant, Suitability for Forecasting or Planner Role then your functionality is old.

Old functionality is painful, but Olde World functionality is a whole different hospital admission queue. If most of the above are pain points, then you are probably severely lagging behind your direct competition. This is more than just painful; it is probably business critical. The opportunity here is that if your competitors are feeling the same pain, now is a chance to get ahead.

Excel Addiction & Swollen Files

If your planning department is addicted to Excel Spreadsheets those files are probably massive! Even so, I bet they still cannot contain all the data that is really required let alone compare to all the previous year's forecasts. How do you know if your forecasts are any good if you cannot compare them? How do you know if planning activities add any value? Perhaps the adjustments made last cycle made the forecast worse.

Data

Data and the information derived from it is at the crux of everything in Demand Planning (and let's face it - everywhere else too). Planning Systems traditionally require a full set of historical data, a method to create a forecast (engine and/or manual entry), sundry additional forecasts (Sales, Marketing, Budget, Incoming Supply etc.) plus conversions, prices, descriptions, attributes all from the Master Data Management of Customer, Product & Organisation. That is a lot of data. See heavy weight above as symbolic of how much it hurts to carry, calculate and process.

Having Master Data Management that is not automatically reflected in Demand Planning Systems is unbelievably sore but not even having Master Data Management at all is really, really, painful for Demand Planners. The data elements below are already covered in the pain points above so for brevity, I have just listed them bullet style.

• Cannot easily Collaborate (collection, sharing and issuing shouldn't be a trial)
• Poor Dashboards (limited functions, not able to share or manipulate)
• Accuracy cannot be measured (no previous forecast or budget or override assessment)
• Master Data issues (New Customers, Products, Categories & Conversions manually entered and not compatible with other systems?)
• No Segmentation (no ability to split out your Products, Customers, Organisations or slice by types of demand)
• No Promotions or Causals (events that impact your forecast should be recorded in the past and the future to help make sense of demand fluctuations)
• No Fiscal Values (Demand should be quantity based but evaluating what to prioritise requires fiscal information.)

Resource

The most important element in Demand Planning is... your Demand Planners! Is this

A. Because without them nothing would appear at the end of the cycle

or

B. Because they add incredible insight and value to the business?

Of course, it's both A and B but if it you are not getting enough B results it will be because they are too focussed on delivering A. Sick systems, feeble functionality, and distressing data will lead to very frustrated planners and lacklustre forecasts!

Workarounds

When Systems and Processes are held together with workarounds to resolve problems the planners will spend time on loading, fixing, consolidating, exporting and importing back tasks instead of actually refining the forecast. Some of the most common reasons to create workaround a pains are:

• Manual Data Loading (not actually connected to the right systems)
• Shallow Hierarchies (you really need at least three levels per Dimension)
• Desired Groupings not available (Using groupings from Finance or Sales may not be suitable for good Demand Planning)
• No Previous Forecasts (this is like running with one leg)
• Budget not available (It will be available, just not in the right place or the right format)

Limiting Planning Maturity Level

Often an undiagnosed malady. Planning systems, processes and people should not remain static. As the world changes and your business evolves, so your planning should grow and develop. Planning maturity cannot happen if the same people perform the same tasks over and over and over again. If those tasks are hampered with system inefficiency and data integrity issues, then it's about survival not evolution.

This isn't brain surgery, everyone needs development. Train up your resources! Still on Excel or using a twenty-year-old system? Never mind, get a plan together, get a strategy for growth. Learn the latest Supply Chain developments, get all trained up & fired up with enthusiasm and get a healthy forecast as a result.

Process

Forecast Cycle is the length of time it takes to create a Forecast from start to finish. Within the forecast cycle there will be many process steps, some will be painful, and they probably prevent efficiency. Cycle time depends on your industry, culture, lead-times, resources, systems, data, habits etc. Typically, the idea is to get more accurate and to get much faster. Sometimes it is best to concentrate on one rather than the other. Process is painful because:

Interminable Forecast Cycle

Everything just takes too long! We all believe that planning faster will result in better flexibility & resilience. All those down-stream activities like production, purchasing, inventory management, logistics etc. will become leaner and faster too. Revenue and margin gain are obtainable objectives, if only we could get faster (and more accurate too).

In the diagram below see where you are in 'Planning As Is' - can you easily get the Forecast Cycle to the proposed 'Planning To Be'? Not being able to get faster is a major irritation but cure is possible with systems and process change.

Re-forecasting

Not being able to change the forecast within the cycle is also a serious discomfort. In an ideal world you should be able to adjust the latest forecast as soon as anything dramatic happens. If fresh insight arrives and the forecast is redundant - why wait until the next cycle (in three weeks!) before sending a fresh, good signal? A good planning solution will enable a fresh signal whenever needed.

Wait! I need to sign that!

If you have no approvals within your process (Sales, Customer, Finance, Management, etc.) then you certainly have a nimble solution. You probably also have audit risks and some poor forecasts. Having no approval may not hurt now but one day it could really, really sting. I recommend a light, exception-based approval touch.

Tick, Tick, Tick, Tick, Tick, Tick, Tick, Tick, Tick, Tick, Tick....

The opposite is even more painful. Too many approvals required from automated emails (triggered by a static value) and spreadsheets with ticks-in-boxes required that nobody wants to do (but have to) create resentment, slavish repetition and potentially, no real checking.

Manual Reports

Those Demand Review reports don't make themselves you know. I mean, literally, they don't. If your planners spend a lot of time doing anything that they'd really rather was automated - it will be having to create and compile the material that is required for the review steps that people don't read let alone show up for.

I guarantee this nasty thing happens a lot: the reports are printed and then a significant, last minute change arrives that needs adjusting in the actual forecast and then... printing again. The waste of time and the wastage of paper, ink and staples is distressing. Automate the creation or better still, perform it in a system where review reports can be specifically constructed.

Opening Old Wounds

If you made it this far, I congratulate you. I hope it wasn't too harrowing.

Painful Consequences

To summarise, the impact of all these problems is forecasts that are not accurate enough, nor created fast enough and have sub-optimal information. Planners that are frustrated with workarounds and executives at having to pay a high price for creaking solutions. The Panaceas to these sores can be found in the form of shiny new technology and re-invigorated processes. They can be obtained from gigantic ERP & Cloud Application providers & Integrators such as SAP, Oracle & Microsoft to dedicated Planning Solution specialists such as AnaPlan, FuturMaster, Optimact, etc.

The solution of choice to drive away pain might be driven by IT Architecture, Business Strategy, Integrator recommendation, ERP Pre-Sales or previous experience. Whichever approach you choose, make sure it resolves the particular challenges that you are facing as well as laying foundations for future change. If you are not sure what the best pain relief might be - look out for the incoming article 'Demand Planning Panaceas'. Coming via a Linkedin post to you soon.